Domain Rules

[ETortoise]

Connecting granaries and temples to the fortifications discussion earlier, what if holdings had two values? One being the holding’s level and the other being it’s breadth. So a citadel is a level 6 castle, but a citadel with a high breadth would mean the domain also had numerous redoubts and smaller fortresses, but a citadel with a breadth of zero would just mean you had the one castle.

Alternatively there could be an infrastructure holding that lets you propagate the effects of your other holdings further.

[Grek]

To that extent, having holdings give a bonus based on the biggest of its type seems the way forward. That is, if you want to improve your piety collection you build the next biggest level of Temple rather than another instance of whatever temple you have right now.

It is important to balance this against the desire for domains to have more than one Temple. While the capital might have a huge cathedral, there should be some reason to build a local church in your smaller settlements.

I think ETortoise has it: base the bonus given by a particular category of building (all piety buildings, all administration buildings, etc) on the single biggest building anywhere in your Domain, but require that a place be within two hexes of any building of that kind in order to benefit. So you end up having one big temple complex in the middle of your capital city or a culturally appropriate holy site, but you also have little regional shrines every 24 miles or so. For a 200 hex domain, that's between ten and fifteen extra temples, depending on how closely your domain approximates a circle.

[Chamomile]

Building zero-level buildings whose purpose is just to expand the area of effect of your capital's big building doesn't have to be difficult at all, either. It can be a straightforward exchange of some fungible resource for a shrine or a guard tower or a magistrate's office, and then you just get to slap one down anywhere in your borders you like.

That said, while keeping track of one such area of effect seems very easy provided you have any kind of editable hex map, keeping track of three seems like it would be much harder to understand at a glance. Given that, it's probably better to just tell your players that fluffwise of course your mega-temple implies the existence of smaller shrines and churches in outlying hexes, but we're not going to model that because it's a massive increase in bookkeeping. If Paris has Notre-Dame, then the village down the way probably has a church or something, and if we really want to we can even have a table that lets you cross-reference the current level of your capital's Administration building with the level of development of the hex you're in right now and see what kind of Administration building this hex implicitly has, but all of that information should be purely "there's a table that tells you how big to describe a market town's church is based on how big the capital's church is, if that sort of thing makes you happy," without any obligation to look it up while actually managing your domain at all.

[Grek]

Having discreet low level buildings with meaningful effects is important for subterfuge actions. What does it mean for your crime syndicate to bribe a sheriff, if you're not marking out his jurisdiction on the map and doing your smuggling where you know you enemy Law and Order bonuses don't apply? When you're trying to culture-flip a town, it should have a definite mechanical impact if you finagle things such that your cousin gets elected as the archbishop of Pelor and makes sermons denouncing the current governor's wickedness, but a considerably lesser advantage if the archbishop is based out of the next town over.

It is also important for warfare. If you have a domain of a hundred hexes, it should obviously have more than one fortress. But it also shouldn't have one in every single hex, because having a fortress every six miles will make the players call bullshit. And because the players will be marching armies around on the hex map and laying siege to places, these fortresses need to be in specific locations rather than being Schrodinger's Battlement. Likewise, if you're a plucky rebel trying to overthrow the lich king, you are probably not going be able to sneak into the Campus Sceleris and stop this month's batch of Vestals from completing the Shadow Merging. But you might be able to set fire to a Hastati training camp, and that should have a measurable impact on the empire's ability to convert manpower into legionnaires, which is based on which camp you burn down and how settled the region around it is.

[Username17]

I feel like design-a-nation needs to be in, although I'm not sold on the term "nation" itself yet. I say this because I definitely think you want room for council hat abilities that let leaders establish tribes/castes/cults/whatevers within their borders that bestow some of the benefits and costs of adding "nations" without having to tack on the flavor of being a pluralistic and racially harmonious society onto every domain that wants to snare phat bonuses. That way the villainous domains have bonuses to acquire and be fractious about even if they're technically all hobgoblins or something.

I'm definitely open to names other than "Nation." I prefer Nation to Race because when Master of Magic talks about "Nomads" and "Barbarians" as races, it's super cringe. But you do want Barbarians and Nomads, and the fact that you can recruit Griffin Riders from the Nomads and Berserkers from the Barbarians is potentially important and the kind of information you'd want to get when asking "who lives in my domain?"

But yeah, while the concept is obvious, English doesn't have a word that precisely means that without coming with a whole lot of weird baggage. Having Hobgoblin "People" or "Community" would be pretty weird, though of course also defensible from a dictionary standpoint.

One thing is that I think it's entirely reasonable for some settings to simply assume that there are multiple races in some of the nations just as some of the races belong in more than one nation. So you could just assume that there were Dryders living in Drow communities and maybe Dryads in Elf communities. Your not-Viking human nation could have some number of Giant-Blooded and Dwarves. Races and Nations don't have to be directly correspondent in either direction.

Alternatively there could be an infrastructure holding that lets you propagate the effects of your other holdings further.

I'd be OK with that. There are sixteen castles in Yorkshire, and some of them are "Fortified Manors" that you probably wouldn't bother with on the domain map and some were already ruins filled with bandits when others were built. But I could definitely see building like eight forts to cover a large domain like York. A fort of some kind covering two dozen hexes would be acceptable, that would imply about two castles for an area the size of Cambridgeshire (41 hexes). And looking it up after making that prediction: there is Kimbolton Castle and Longthorpe Tower, which seems about right. One fort to give the fort bonus to 24 hexes and a smaller fort to extend the bonus to the rest of the domain.

Seems workable.

-Username17

[Grek]

Perhaps a bit dry, but Demographic? Elves are a demographic, barbarians are a demographic, cultists of Nemelex Xobeh are a demographic, amazons are a demographic, etc.

[Username17]

Building on the idea of nations having more than just a single race in them, I think the model has to be Heroes of Might and Magic. Although with a substantially larger cast of nations because it's a pen and paper game and adding content is very easy - and also because every Heroes Game has had substantially less factional content than it should. Every time they bring out a new entry in that series with less than a dozen faction types I'm like "What the actual fuck do you think you are doing?"

Anyway, what this corresponds to is that for every nation you have a basic progression of military units ([$Guy] with Spear, [$Guy] with Armor and Sword, and so on and so on), and also that you unlock special slow to recruit units with certain levels of development. And here I think is where we turn our eyes to Master of Magic for insight. That the holding "Fantastic Stables" lets you recruit Pegasus Riders in areas with Elves and Griffin Riders in areas with Nomads. With more nations and deeper troop rosters you can do similar things - the nation of Forest Kith Goblins can recruit Trolls when it has the upgrade that also allows the nation of not-Vikings to recruit Frost Giants.

Each nation doesn't have to get options for all the slots, and it's probably better if they don't. The Drow might not have a Giant and the Troglodytes might not have any War Machines.

Special unit types could include:

• Flying Beast (Doom Bat)
• Flying Rider (Pegasus Rider)
• Giant (Troll)
• Giant, Elite (War Troll)
• Magical Person (Satyr)
• Monster (Hydra)
• Monster Rider (Wyvern Rider)
• Outsider (Genie)
• War Dog (Wolf)
• War Dog, Magical (Blink Dog)
• War Machine (Cannon)
• Wizard (Wizard)

So if you have a big temple in the middle of your Dwarf Fortress, you can spend your Piety on recruiting those Dwarf Ancestors instead of doing festival actions to summon regular angels.

-Username17

[OgreBattle]

How would you feel about changing the word 'piety' to 'merit'?

Merit can be used to mean doing something "to have a claim to a future reward from a graceful God.

'piety' makes me think more of love and respect for a divinity than accumulating premium divine currency to spend

[DrPraetor]

There needs to be a scale transition where you care about having 40 or 70 tiny men with spears, to where you only care how many zounds you have.

I'm not sure how it should work mechanically, but in a tabletop RPG you can't be deciding how to split patrol duties among 75 platoons of kobold spearmen, it just needs to be some policy you set. At the same time, at a lower level/scale, if your one platoon of kobold spearmen is half-strength or well-equipped or whatever needs to matter.

[Username17]

How would you feel about changing the word 'piety' to 'merit'?

Merit can be used to mean doing something "to have a claim to a future reward from a graceful God.

'piety' makes me think more of love and respect for a divinity than accumulating premium divine currency to spend

I actually am fine with "Merit." I only broached "Piety" because it's the Crusader Kings concept. Although in Crusader Kings the Hindus get "Karma" instead, which might be better still as it implies something that can specifically be used up. Not everything has to be pseudo-English, and Karma or Kismet are English words now and probably better names for a spiritual currency.

Although it would be possible to split things up such that the spiritual currency of the different pantheons had different names. So you could maintain a tally of Karma with one pantheon, of Kismet with another pantheon, Destiny with a third pantheon, Arete with another pantheon, and Doom with the pantheon of gratuitous villainy. That might be too fiddly, especially as most of the time domains are only going to permit one pantheon to operate and they could just all use Karma.

There needs to be a scale transition where you care about having 40 or 70 tiny men with spears, to where you only care how many zounds you have.

I'm not sure how it should work mechanically, but in a tabletop RPG you can't be deciding how to split patrol duties among 75 platoons of kobold spearmen, it just needs to be some policy you set. At the same time, at a lower level/scale, if your one platoon of kobold spearmen is half-strength or well-equipped or whatever needs to matter.

As it plugs into an RPG, things can go the personal scale when there's just 40 spearmen. There is a penumbra where things get numerous enough that it's a problem to go up and personal but also that there aren't enough to cover hexes and stuff. I'm hoping that by interacting with the council part, that you can keep the RPG bits relevant by scaling them up as domains and economies and armies increase in size. You have RP sessions where you talk to burghers that represent larger civil interest groups as the economies grow.

-Username17

[jt]

I did some research into hex and counter wargames when thinking about mass combat rules. They have these scales (nomenclature is not consistent):
[*] Man to man: What we're used to in D&D land, one counter is one soldier. Rare.
[*] Tactical: A counter represents a bunch of similar soldiers (similar as in all infantry or all cavalry, there could be support mixed in). The counter can be anything from a squad to a division. Most similar to HOMM.
[*] Operational: A counter represents an entire army. A battle is one roll.
[*] Strategic: Like operational, but so zoomed out that there's also an economic game attached.

I think the tactical/operational split here is useful to apply in an RPG setting. At the tactical scale, you can have an army of 40 pegasus knights, 200 dwarven heavy infantry, a mage circle, and three units of 500 regular infantry, and you have a pretty good idea how to drop the PCs into that. At the operational scale, that's just a strength 9 army, and you'd have to convert to tactical scale before even knowing what's in it.

I like Grek/ETortoise's idea for holdings, where you actually design your best castle, and then copy its benefits by putting generic castles in hexes. When you zoom in on that and need to figure out what a generic castle looks like, it looks like something designed by somebody who was heavily influenced by the castle in the capital. Not only is that convenient, it's realistic - same as how almost every subway system in former Communist Bloc countries is the Soviet Triangle copied from St. Petersburg.

I think this same trick can be used for a tactical/operational split in army management. You actually put some thought into the grand central army of the domain. Anything else is just a strength X army. And when you go need to go zoom in on those other armies, they look like somebody trying to copy the central one.

[Username17]

I think this same trick can be used for a tactical/operational split in army management. You actually put some thought into the grand central army of the domain. Anything else is just a strength X army. And when you go need to go zoom in on those other armies, they look like somebody trying to copy the central one.

That is of course the core of it. If your main army is mostly Dwarven Heavy Infantry and Goblin Warg Riders, the peripheral armies can be that if you ever have to zoom in on them. If your army uses Displacer Beasts as skirmishers, then the peripheral armies can do that too.

Where it gets complicated is the really special military assets. If your army has a Black Dragon or a Pit Fiend it just has the one. You don't get copy pasta versions of those in your border garrisons.

-Username17

[deaddmwalking]

You don't get copy pasta versions of those in your border garrisons.

I'm a little torn on this. If your provincial armies reflect your core army, it's true you don't have your black dragon stationed on the northern border, but it's also not impossible to consider that it could make a strafing run/minor contribution.

Ie, if you have a key special unit, and you're at an operational split, you don't have to KNOW whether the dragon is there or not - whatever contribution it makes is just factored in.

If you're playing at the Tactical Level redeploying a special unit is a big deal. There's a good episode of Young Indiana Jones Chronicles where the Germans are moving up two rail-mounted howitzers (Big Berthas) which I think in this context qualify as special units (there weren't a lot of them and their presence was a big deal). That would work out well for a PC-led action.

So at the Operational Level, border garrisons aren't as strong as the central army, and to the degree that they reflect the central army it could be tiny black dragons or occasional visits by the big black dragon but ultimately you don't care because at that level you're not zoomed in enough to see individual troops.

[Thaluikhain]

Do different scales overlap? Are there battles you can run as big examples of the small scale or as small examples of the big (hopefully with the same results)?

Or do you, say, have 10 a side or less battles and 100 a side or more battles, but for some reason never anything in betwee?

[Username17]

There has got to be a point where special units become generalized. Like, there's a point at which you have 'A Harpy' and there's a point at which you have 'Some Harpies' but there's a whole other point at which Harpies are simply part of your army and can be expected to be available in otherwise unspecified military units.

The key insight I think is the same with the fortifications. That once you build access to something it requires additional relay stations to spread it throughout a large domain. A small domain with just a few hundred soldiers doesn't have to bother - the main army is also the entire army so however many spider demons you have in it is just how many spider demons there are.

Once your army is big enough that you might need to zoom in on what 'previously unnamed military force #5' is doing, that's when you need to build other stuff to propagate elite and specialist units into those forces when you zoom in on them. Otherwise your main force gets all the weird elite stuff and all the place holder forces are like the equivalent of Dominions 5 Province Defense.

So let's same your Nation (Demographic? Identity? Folk? Populace?) are Amazons. If you get big enough to have multiple forces, your extra forces are going to be women with bows and women with spears. And if you recruit women on Pegasi or Harpies and Medusae and Lamiae into your main force, the extra forces are still just women with bows and women with spears. But if you recruit Harpies into your main force and then also build skynest outposts, then your extra forces will also have Harpies.

And this applies even to non-fantastic troops. If your peoples are Halflings then you get a bunch of slingers and militia in your additional peripherals. And it's going to take a stables network to even have Pony Riders spawn in those things.

-Username17

[Kraydak]

I think that deck-building is a mechanic worth exploring for domain rules:
1) You get natural scaling via deck-thinning (e.g. remove 3 towns from the deck to build a city)
2) You get natural random events to respond to (e.g. this month we drew "Boring Badlands Orc Raiders")
3) You get a lot of flexibility in how things get modeled (e.g. after mostly clearing the Boring Badlands the DM adds "Doughty Independent Peasants" for the new inhabitants, an "Uncleared Ruins" because no one has time for everything, an "Orc Raiders" and an "Unfriendly Elven Kingdom" from across the borders to the deck). Independent or sponsored adventuring parties become cards.

This would definitely require a tutor-type Domain action, of course. It could be an action-resolution layer on top of a more careful Koku-based accounting system for crunchier people, or the entire thing in a more Euro-gamey iteration.

[jt]

Worth The Candle uses "Polity" as their generic nation-word. It's grown on me.

You don't get copy pasta versions of those in your border garrisons.

Going from a more abstract representation to a less abstract representation requires making shit up, so this is a place where part of the rules are have to be just ideas and guidelines for the GM.

Determining The Contents Of Other Armies
Armies besides the central army usually just have a strength number and not any specific units, but sometimes it's necessary to find out what those units are. Start with a copy of the central army, and start removing pieces of it to bring it down to the other army's strength.
[*] The central army will often have unique units that can't be copied - say the only black dragon in the domain. These are usually the first thing you should remove.
[*] Unique units may also be replaced with lesser substitutes. The commanders of a border garrison know that they can't get a black dragon, but are still following a military doctrine that assumes they have one, so they manage to tame a wyvern.
[*] Consider the origin of your special units and prefer removing ones far from home. Your northern provinces may drop their spider-riders first while your southern provinces drop their dwarven infantry first.
[*] It's usually best to remove more special units than core army units, but sometimes an army might be just the specialists. This pairs well with out of the way places and defensible positions.
[*] As a domain gets larger, it becomes more likely that something interesting has happened that the players never heard about. A secondary army might have persuaded another dragon to fight with them, a small border town might have a friendly iron golem as part of its garrison. The players are likely to want to know how this happened, and the story behind it can introduce new minor characters and plot hooks.

Do different scales overlap? Are there battles you can run as big examples of the small scale or as small examples of the big (hopefully with the same results)?

With good minion rules, I think individual combat can comfortably scale up to ~5 PCs vs 20 grunts, and uncomfortably up to twice that. Say a unit in tactical combat is 4 to 5000 soldiers, and it can scale from ~12 to ~100k per side. Operational scale combat can scale from one to a zillion, since it's basically an opposed, "How much are we feeding these people? + How special are they?" roll.

A good system can get the same results on average, but only across whatever army compositions it considers equivalent. Tactical scale is going to say that 20 level 1 human fighters vs 20 level 1 human fighters is a tossup, even if your individual combat rules say that 20 spiked chain trip fighters vs 20 two weapon fighters is a complete shutout. Operational scale boils entire armies down to a strength number, and strength 3 vs strength 3 is a tossup, even if that's 20 ents vs 100 pyromancers. Operational scale especially should have less stupid results the bigger of a battle you use it on - ents vs pyromancers is less of a problem if they're each one unit in an army with a dozen.

[Username17]

Dominions calls the nations various fantasy names like Vanheim and Themiskyra rather than descriptive terms like "Vikings" or "Amazons." And for a completed game, I think that's right. So when we talk about "Orcs" or "Lizardfolk" those are playtest placeholder names. The actual polities would have fantasy names.

But it's also important that the races and monsters won't just appear in one polity. You have your Not-Scottish Dwarves and your Not-Sumerian Dwarves as polities. And you have Harpies that show up in the Amazon polity and Harpies that show up in the Beastman polity and those units can be literally exactly the same. Further, at the level of abstraction of the military game, an Amazon riding on a Pegasus and an Elf riding on a Pegasus can just be the same "Pegasus Rider" unit.

Because it's a pencil and paper game, adding polity content is extremely easy, and it can support a lot of things.

Amazons

• Harpy
• Nymph
• Pegasus Rider
• Medusa
• Lamia
• Sphynx

Mountain Dwarves (the OG D&D Dwarves)

• Cave Bear
• Doom Bat
• Wolf
• Goat Rider
• Stone Giant
• Golem

Orcs

• Dire Boar
• Wyvern Rider
• Ogre
• Ettin
• Wolf
• Doom Bat

Forest Goblins

• Wolf
• Warg Rider
• Giant Beetle
• Troll
• Hydra
• Satyr

Steppe Goblins

• Wolf
• Warg Rider
• Hill Giant
• Manticore Rider
• Lamia
• Giant Bee

Halflings

• Dire Boar
• Cockatrice
• Giant Goat
• Eagle Rider
• Gorgon
• Treefolk

Drow

• Giant Spider
• Dryder
• Phase Spider
• Shadow Beast
• Wisp
• Gargoyle

-Username17

[Username17]

Not of course that every polity must have 6 creatures that aren't '[$Guy] with [$Weapon]', but I think it is clarifying.

Elves (The Forest Kind)

• Dryad
• Pixie
• Pegasus Rider
• Giant Owl
• Unicorn Rider
• Treefolk

Deep Dwarves (Duergar)

• Hell Hound
• Giant Spider
• Doom Bat
• Fire Giant
• Salamander
• Gargoyle

Nordic Dwarves

• Winter Wolf
• Dire Wolf
• Bear
• Frost Giant
• Boar Rider
• Valkyrie

Shadow Dwarves (Derro)

• Giant Spider
• Doom Bat
• Basilisk
• Shadow Beast
• Golem
• Naga

Sumerian Dwarves (Zharr)

• Gian Scorpin
• Gorgon
• Ogre
• Griffin Rider
• Sirrush
• Shedu

-Username17

[Thaluikhain]

Should each have monsters from set categories (unless not having so to be different)? So most of them have a big monster like an ogre, some sort of flying mook, some some of elite magic cavalry?

[Username17]

Should each have monsters from set categories (unless not having so to be different)? So most of them have a big monster like an ogre, some sort of flying mook, some some of elite magic cavalry?

The more you care about 'game balance' and the less work you want to do, the more likely you are to answer 'Yes.'

For the purposes of a board game or a computer simulation of a board game where one player might be the Chaos Dwarves and another player might be the High Elves and so on and you expected a reasonably level playing field between them, it's very difficult to balance things without all the players having access to the same kinds of troops. Balancing 'Lamasu are stronger and more expensive than Giant Owls' is challenging, but nowhere near as challenging as balancing 'Gorgons are slow moving, high defense cow monsters and Giant Owls are stealthy flying skirmishers.'

For an RPG extension, the various groups don't have to be balanced because the positions aren't balanced. The players are going to be doing zero to hero stuff and their heroic actions are going to include taking an 'underdog' domain to great heights through the use of character abilities and trickery. It's OK if the weasel riders of the Kobolds are just worse than the Warg Riders of the Goblins, because the players are explicitly looking to improve their Domains and that requires both a place to improve from and a place to improve to.

Now a caveat to that is that you don't want to lock players out of portions of the game altogether. So it might be fine for Troglodytes to have no cavalry and no airforce, but only if you don't actually expect the players to have an exclusively Troglodyte domain at any point. Maybe the Grimlocks literally have no archers, but again that's only OK because the player characters are not expected to be chiefs of Grimlockia and if that's the campaign it was chosen with the limitations in mind. Whatever main parts of the game are, the polities you expect the PCs to rule over should be able to interact with it in some way. The Mountain Dwarves may not have any light infantry, but they'll have infantry of some kind.

I would suggest that every presumptive player facing polity have access to:

• Flyers
• Infantry
• Cavalry
• Archers
• Monsters

And sometimes you have stuff like Pegasus Riders that are Flying Archers and sometimes you have stuff like Fire Giants who are basically Monsters who are also Infantry. And sometimes you have stuff like Wyverns that are Monsters and also Flyers.

-Username17

[deaddmwalking]

I would suggest that every presumptive player facing polity have access to:

• Flyers
• Infantry
• Cavalry
• Archers
• Monsters

I think this is one of the situations where what you don't have is part of what makes a faction feel unique more than what they do have. If dwarves have no or relatively bad flyers, but they have the best infantry, that's something.

Just for the sake of example I'm going to assume that each unit has a Strength/Toughness like a Magic the Gathering Card.

Dwarves might have Infantry 4/4 while the elf infantry is 2/2, but Dwarves might have Bear Cavalry 5/3 while the Elves have 5/2 Flying.

[merxa]

Where are the battle mages in that list?

Having a unit that casts fireball, or having an embedded skirmisher in every unit that can cast fireball seems important to model since the deployment of magic is the obvious must be fixed issue for fantasy warfare.

[Username17]

Where are the battle mages in that list?

Having a unit that casts fireball, or having an embedded skirmisher in every unit that can cast fireball seems important to model since the deployment of magic is the obvious must be fixed issue for fantasy warfare.

Battle mages have to exist, obviously. But I don't think they are a question of polity. That is, whether you get Necromancers or Geomancers is a separate question to whether you have Elves or Dwarves.

If you build a Tower of Bones, you can get Necromancers for your army. But you can build that tower in a Halfling shire if that's what you want to do. Or you could build one in a Bullywug marsh. Either way, the choice to build any particular kind of mage school is polity independent, although I can well imagine that there might be some council hats that would give you a discount on going for one or another. Like, being a Master Gardener might give you a discount on building a Druid Henge or Geometer's Forum. Being Court Necromancer presumably gives you a discount on making a Tower of Bones or a Vault of Contracts.

Now the actual divisions of magic are wholly arbitrary. You could divide things up by four or five elements, you could divide things up by the seven forbidden arts or the eight immortals or fucking whatever. In this case, I'm looking to merge this with Kitchen Sink Fantasy, which means you get:

• Druid
• Elementalist
• Enchanter
• Illusionist
• Necromancer
• Paladin
• Psion
• Warlock
• Wizard

Which is nine, and I admit that's kind of a lot. But I also figure that most PC domains will probably end up with two to five flavors of magician in their army.

-Username17

[Emerald]

That is, whether you get Necromancers or Geomancers is a separate question to whether you have Elves or Dwarves.

If you build a Tower of Bones, you can get Necromancers for your army. But you can build that tower in a Halfling shire if that's what you want to do. Or you could build one in a Bullywug marsh. Either way, the choice to build any particular kind of mage school is polity independent, although I can well imagine that there might be some council hats that would give you a discount on going for one or another. Like, being a Master Gardener might give you a discount on building a Druid Henge or Geometer's Forum. Being Court Necromancer presumably gives you a discount on making a Tower of Bones or a Vault of Contracts.

It might also make sense to tie that to terrain type or major landmarks, such that you get discounts on Druid Henges built in large forests, Pyromancer Towers built in volcanoes, and so forth; the same could go for other structures (Pegasus Stables are cheaper in forests and mountains, Winter Wolf Dens are cheaper in tundras, etc.) if you weren't planning on that already. That would tend to lead to the kind of stereotypical associations people would lean toward anyway (wood elves having lots of druids, mountain dwarves having lots of earth mages, drow having necromancers, etc.) without enforcing that in any way and without requiring you to write up a bunch of racial modifiers for that.

Also, having discounts for roles plus discounts for terrain kinda emergently gives you the "themed domains" Dean was talking about a few pages back. If you really want to go all-in on the "forest" theme, you can build all of your structures in forests and have all the players pick forest-related roles like Master Gardener/Ranger Lord/Hierophant/etc. so you're really good at producing forest-y structures and units, at the cost of sacrificing thematic and mechanical diversity--and, again, without having to write up specific rules to support the concept.